HAL author manuscript - HAL-Inserm

HAL author manuscript Digestive Diseases and Sciences 2006;52(1):84-92

The original publication is available at www.springerlink.com

Effect of rebamipide on the colonic barrier in interleukin 10 deficient mice HAL author manuscript

David Laharie, MD,1, Sandrine Ménard, PhD,2, Corinne Asencio1, Teresita Vidal-Martinez, MD.2, Anne Rullier, MD3, Frank Zerbib, MD, PhD, 1, Céline Candalh2, Francis Mégraud, MD1, Martine Heyman PhD.2

inserm-00211574, version 1

and Tamara Matysiak-Budnik, MD, PhD2

1

INSERM-ERI 10, Université Victor Segalen, Bordeaux, France

2

INSERM EMI0212, Faculté Necker- René Descartes, France

3

Service d’Anatomo-pathologie, Hôpital Pellegrin, Bordeaux, France

Running title: Rebamipide and colonic barrier

Address for correspondence: Tamara Matysiak-Budnik INSERM EMI-0212 Faculté Cochin Necker 56 rue de Vaugirard 75730 Paris, France Tel: 33 1 40 61 56 33 Fax: 33 1 40 61 56 38 Email:[email protected]

1

Acknowledgement: This work was supported by Otsuka Pharmaceutical, Tokyo, Japan.

HAL author manuscript

ABSTRACT Our aim was to study the effect of a mucosal protective agent, rebamipide, on the colonic barrier and the immune response in colitis-prone interleukin-10 deficient (IL-10-/-) C57BL/6 mice infected with Helicobacter hepaticus. After sacrifice, in all mice, control, or previously


infected with H. hepaticus, or previously infected and treated with rebamipide enema, a histological examination of colonic samples was performed, intestinal permeability was studied in Ussing chamber, and mesenteric lymph node proliferation and cytokine secretion were measured. Mice treated with rebamipide, presented a reinforcement of the distal colonic epithelial barrier, an increase of mesenteric lymph node cells proliferation and of IFN and IL-12 secretion. These results indicate that in IL-10-/- mice with mild colitis, rectally administered rebamipide reinforces the distal colonic barrier and has a slight Th1 immunostimulatory effect on mesenteric lymph node cells. These properties could be helpful in the management of some inflammatory bowel diseases.

Keywords: rebamipide, intestinal permeability, colitis, interleukin 10, immune response

2

HAL author manuscript

Inflammatory bowel diseases (IBD), which include Crohn’s disease (CD) and ulcerative colitis (UC), have multifactorial etiologies involving genetic, environmental, and immunological factors. One of the constant features in IBD is the increase in intestinal permeability (1, 2). Under normal conditions, the intestinal epithelium forms a barrier that


simultaneously regulates the absorption of nutrients and water, and prevents the uptake of bacteria and bacterial products from the intestinal lumen. An impaired intestinal barrier function is observed in patients with CD (3, 4) but also in healthy relatives of these patients (5). The unrestricted passage of bacterial antigens from the intestinal lumen may induce the chronic activation of mucosal immune cells, a critical component of the inflammatory process associated with IBD. Current treatment of IBD have focused on drugs able to decrease inflammation but the question of limiting antigen access to the submucosa by reinforcing the epithelial barrier has not been addressed. Rebamipide is a mucosal protective agent which, in addition to stimulating the production of endogenous prostaglandins in the gastric mucosa and improving ulcer healing (6), exerts a positive effect on the gastric barrier function. It reinforces the barrier’s integrity in basal and inflammatory conditions and it inhibits the increase in macromolecular transepithelial transport induced by Helicobacter pylori (7, 8). Experimental data have shown that rebamipide can prevent dextran sulfate sodium-induced colitis in rats (9). In addition, in a human case report, rebamipide exhibited a beneficial effect on a proctitis type of ulcerative colitis (10). Murine models of enterocolitis mimicking IBD, have been described (11, 12). Interleukin-10 deficient (IL-10 -/-) mice infected with H. hepaticus have been shown to develop a moderate

3

to severe inflammation of the large bowel (13). We hypothesised that in this model of colitis the reinforcement of barrier function and modulation of the intestinal mucosal immune system responses to luminal antigens would be a good therapeutic approach. Therefore, our aim was HAL author manuscript

to study the effect of rebamipide on intestinal permeability and immune response in IL-10-/mice infected with H. hepaticus.


4

METHODS Experimental protocol Forty female IL-10-deficient C57BL/6 mice (B6.129P2-IL-10tm1Cgm/Crl) obtained from HAL author manuscript

Charles River Laboratories, L’Arbresle, France (aged 5 weeks at the beginning of the experimental protocol) were used in this study. They were housed 12 per cage, at 20°C under specific pathogen free conditions during two weeks and then shifted to conventional litter and to a standard non-sterile diet.


Altogether, 4 groups of mice were studied: Group I = Control group (n = 12): IL-10deficient non-infected mice, Group II = Infected group (n=12): IL-10-deficient mice infected with H. hepaticus (strain CCUG 33637, ATCC 51448) by three oral gavages with 2 x 10 9 CFU in 500 µl at the age of 5 weeks, Group III = Rebamipide group (n=12): IL-10-deficient mice infected with H. hepaticus and then, starting from one week after infection, treated daily with rebamipide enema (300 µg of rebamipide in 100 µl of carboxy-methyl cellulose used as a vehicle) during 9 weeks; Group IV = Placebo group (n=4): IL-10-deficient infected mice, receiving rectally 100 µl of the vehicle carboxy-methyl cellulose during 9 weeks instead of rebamipide enema. H. hepaticus infection was confirmed by semi-nested PCR in stools. After DNA extraction from the fecal samples using the QIAamp

DNA stool mini kit (Qiagen) according to the

manufacturer’s instructions, a PCR was performed using primers F2 16S and R4 16S, detecting conserved bacterial 16S ribosomal DNA(14). These two oligonucleotides, produced an amplified product of 1490 pb. 2.5 l of the fecal DNA preparation was added to a 25 l (final volume) reaction mixture containing 1X reaction buffer (with 15 mM MgCl2), 0.5 M each of the two primers, 200 M each deoxynucleotide, 0.1 g of bovine serum albumin per l and 0.025 U/ l of Taq polymerase (Promega, Charbonnières les Bains, France). The following conditions were used for amplification: an initial denaturation at 94°C for 5 min

5

was followed by 40 cycles (denaturation at 94°C for 30s, annealing at 55°C for 30s, elongation at 72°C for 1min20s) and a final elongation at 72°C for 7 min. After purification with SephacrylTM S-400 high resolution (Amersham Biosciences, Saclay, France) column, a HAL author manuscript

second set of primers was used for the amplification of a Helicobacter spp specific 16S ribosomal DNA fragment. These two oligonucleotides, 5’GCTATGACGGGTATCC3’ (C97) and 5’GATTTTACCCCTACACCA3’ (C98), produced an amplified product of 398 pb. The reaction mixture was the same as this used for the first amplification but without bovine


serum albumin. After a denaturation at 94°C for 5 min, 40 cycles (denaturation at 94°C for 1 min, annealing at 54°C for 1 min and elongation at 72°C for 30s) and a last elongation at 72°C for 7 min were performed. PCR was still positive in all infected mice 4 weeks after infection. Throughout the protocol, mice were weighed twice a week and their general behaviour was recorded. Experiments were conducted in accordance with the ethical guidelines of the French Veterinary Department. Clinical and histological examination At the end of the treatment protocol, the mice (aged 16 weeks) were sacrificed, and MLN and the entire colons were obtained. Proximal (caecum) and distal colonic samples were fixed in 4% formaldehyde for histologic examination while other fragments were opened along the mesenteric border and used to study permeability in Ussing chambers. Microscopic score Fixed tissues were embedded in paraffin, sectioned at 5 µm, and stained with hematoxylin and eosin. Sections were examined blindly by a single pathologist (A.R.). Results were expressed using a histological index ranging from 0 to 4 as previously described (15). This index is based on the degree of epithelial layer erosion, goblet cell depletion, and inflammatory cell

6

infiltrate [0=normal; 1=minimal evidence of inflammatory infiltrate; 2=significant evidence of inflammatory infiltrate (cryptitis, crypt abscesses); 3=significant evidence of inflammatory infiltrate with goblet cell depletion; 4=significant evidence of inflammatory infiltrate with HAL author manuscript

erosion of the mucosa].

Epithelial barrier of the colonic mucosa One caecal fragment and one distal colonic fragment from each mouse were laid on a


millipore filter (8 µm pores) and mounted as flat sheets in Ussing chambers with an exposed surface area of 0.2 cm2. They were bathed on both sides with 1.5 mL of Ringer solution containing 5 mM mannitol, which was continuously thermostated, circulated, oxygenated and maintained at pH 7.4 with 5% CO2 / 95% O2. The mucosal and serosal bathing solutions were connected via agar bridges to calomel electrodes for measurement of the transepithelial potential difference (PD) and to Ag-AgCl electrodes for current ( I) application. The tissues were kept under open-circuit conditions and electrical measurements were performed using a DVC 1000 voltage/current clamp (World Precision Instruments, Aston, U.K.). The tissues were regularly clamped at 1 mV to measure the I and calculate the electrical resistance (R) according to Ohm’s law (PD=R x I). Horseradish peroxidase (HRP, MW 40 kDa) was used as a soluble protein marker of the transcellular transport pathway (transcytosis). Mannitol was used as a small molecular tracer (MW 182 daltons) of the paracellular pathway. HRP (0.4 mg/ml), 3H-HRP (37 kBq/ml) and 14

C-mannitol (12.2 kBq/ml) were simultaneously added to the mucosal compartment bathing

intestinal fragments. Samples (800 µl) were taken from the serosal compartment at 10, 30, 50, 70, 90 and 110 min and replaced by fresh Ringer solution. Intact HRP fluxes from the mucosal to the serosal compartment (JHRPi) were determined by enzymatic assay (16) on 200 µl of serosal samples. Total HRP fluxes (intact + degraded = 3H-equivalent HRP fluxes) and

7

14

C-mannitol fluxes were assessed on 500 µl serosal samples by counting the 3H and

14

C

radioactivity respectively, using liquid scintillation photometry (Kontron, betamatic) after double-labelling correction. Degraded HRP fluxes were calculated as 3H-equivalent HRP HAL author manuscript

fluxes minus intact HRP fluxes. Mean steady-state fluxes obtained from 50 to 110 min are presented in figure 3.

Activation status of mesenteric lymph node cells


Mesenteric lymph nodes were collected at sacrifice and single cell suspensions were prepared in Cerottini culture medium (DMEM glutamax supplemented with 8% heat-inactivated fetal calf serum, asparagine 36 mg/L, arginine 116 mg/L, folic acid 10mg/L, HEPES 1g/L,

-

mercaptoethanol 0.05 mM, penicillin 100 U/ml, streptomycin 100 µg/ml, fungizone 1 µg/ml). Generally, the total cell count was about 20 x 106 cells/mouse which were adjusted to 2 x 106 cells /ml in the culture medium. Lymphocyte stimulation test: 3H-thymidine incorporation Basal and stimulated cell proliferation was analyzed using 3H-thymidine incorporation. Cells were seeded in 96-well culture microplates at 4 x 105 cells/well (180 µl) and were stimulated or not in duplicate with 20 µl of soluble extract of E. coli or H. hepaticus, at the final concentration of 1 and 3 µg/ml, respectively. Soluble E. coli extract was prepared from cultures of E. coli isolated from human flora. The bacteria were harvested and washed thoroughly in PBS, and their lysis was performed by sonication at 4°C. Cell debris were removed by centrifugation at 8,000 x g for 30 min at 4°C, and the supernatant was sterile filtered. Its protein content, measured by the Bradford assay (Pierce, France), was 1.2 mg/ml. LPS concentration of this extract was measured using the Limulus amebocyte lysate (Charles River Endosafe, Charleston, VA, USA). The soluble extract contained 8.6 µg/ml of LPS, leading to the final concentration of 86 ng/ml in culture

8

wells. Soluble H. hepaticus extract was prepared from H. hepaticus strain cultures according to the same protocol. There was no detectable LPS in this extract (< 80 pg/ml). After a 72 hour incubation in the presence of soluble E. coli or H. hepaticus, [methyl-3H] thymidine (15 HAL author manuscript

kBq/well) was added. 3H-labelled DNA was extracted 18 hours later using a cell harvester (Inotech, Perkin-Elmer). Filters were counted using a 1450 Microbeta Wallac Trilux microplate reader (Perkin-Elmer Life Sciences, Paris, France). Results are expressed both as counts per minute (cpm)/well and as stimulation index (SI), i.e. cpm ratio (stimulated cells /


non-stimulated cells). Cytokine secretion: ELISA and flow cytometry ELISA: MLN cells were also seeded on 24-well culture plates (approximately 2 x 106 cells/well) in 1 ml of Cerottini culture medium with and without stimulation by soluble E. coli (1 µg/ml), and H. hepaticus (1 µg/ml) in duplicate. After 72 hours, supernatants were collected and frozen at –20°C until assayed. Four cytokines were assayed: IFN

TNF , IL4

and IL12 using duoset ELISA kits (R&D, Abingdon, UK).

Flow cytometry: IFN secretion by MLN cells in basal conditions without stimulation was assessed by flow cytometry using a mouse IFN secretion assay (Miltenyi Biotec). MLN samples were seeded on 24-well culture plates (106 cells/well) in 1 ml of RPMI containing 5% mouse serum for 16 hours. MLN cells were washed in cold buffer, capture antibody against IFN /CD45 was added for 5 min on ice, then the samples were incubated for 45 min at 37°C under slow continuous rotation to allow cytokine secretion. Samples were washed and treated with anti-IFN APC-conjugated detection antibodies for 10 min on ice. The cells were counterstained with monoclonal antibody against CD3-PE, CD4-FITC or CD3-PE, CD8FITC.

Activation markers on T cells 9

CD44 and CD62L markers were used to measure T cell activation. In activated cells, CD44 expression is increased and CD62L expression is decreased. MLN cells were stained with anti-CD3-PE and anti-CD44-FITC antibodies (Becton Dickinson, Le pont de Claix, France). HAL author manuscript

CD62L staining was done using a biotinylated anti-CD62L antibody which was detected by a PE-cyanin7- Streptavidin conjugate (Becton Dickinson).


Statistical analysis Statistical analysis was performed using the SAS package. The results are expressed as mean SD and comparison of different parameters among the groups was performed by using analysis of variance. The general linear model procedure was also used for multiple group-togroup comparisons. Because of a small number of mice in placebo group, comparisons with this group were performed using non-parametric tests (Wilcoxon and Kruskal-Wallis). The differences were considered significant for p